econstor
A Service of
zbw
Make Your Publications Visible.
Leibniz-Informationszentrum
Wirtschaft
Leibniz Information Centre
for Economics
Thévenot, Laurent; Jagd, Søren
Article
The french convention school and the coordination
of economic action: Laurent Thévenot interviewed by
Søren Jagd at the EHESS Paris
Economic Sociology: European Electronic Newsletter
Provided in Cooperation with:
Max Planck Institute for the Study of Societies (MPIfG), Cologne
Suggested Citation: Thévenot, Laurent; Jagd, Søren (2004) : The french convention school
and the coordination of economic action: Laurent Thévenot interviewed by Søren Jagd at the
EHESS Paris, Economic Sociology: European Electronic Newsletter, ISSN 1871-3351, Vol. 5,
Iss. 3, pp. 10-16
This Version is available at:
http://hdl.handle.net/10419/155834
Standard-Nutzungsbedingungen:
Terms of use:
Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen
Zwecken und zum Privatgebrauch gespeichert und kopiert werden.
Documents in EconStor may be saved and copied for your
personal and scholarly purposes.
Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle
Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich
machen, vertreiben oder anderweitig nutzen.
You are not to copy documents for public or commercial
purposes, to exhibit the documents publicly, to make them
publicly available on the internet, or to distribute or otherwise
use the documents in public.
Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen
(insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten,
gelten abweichend von diesen Nutzungsbedingungen die in der dort
genannten Lizenz gewährten Nutzungsrechte.
www.econstor.eu
If the documents have been made available under an Open
Content Licence (especially Creative Commons Licences), you
may exercise further usage rights as specified in the indicated
licence.
10
THE FRENCH CONVENTION SCHOOL AND THE COORDINATION OF
ECONOMIC ACTION.
Laurent Thévenot Interviewed by Søren Jagd2 at the EHESS Paris.
Starting in economics very early you seem to have had an interest in a sociological
perspective on economic action. Where did this sociological inspiration come from?
Several of the people who participate in the convention tradition have the same background
that I have, the école polytechnique, which is based on mathematics and physics. It is the
highest level within an Ecole Normale Superieure on this topic. It means that all these people
have their first training in economics. In France it is not possible to enter this kind of school
with both an interest in humanities and in mathematics. At some step in the former training
you have to choose. Why did I say that? Just to say that many of the people who find
themselves within this school are actually quite interested in other areas. This was clearly the
case of Robert Boyer of the regulation school, Robert Salais, André Orléan, Jean-Pierre
Dupuy, and me. The five of us are from the école polytechnique and we are all interested in
other things than mathematics.
After the école polytechnique the best students chose to make an application for a second
grande école for their Master degree. I chose INSAE, which leads to INSEE (The National
Institute for Statistics and Economic Studies). You need to know that in France INSEE is not
only an economic bureau. In the same institution you have people producing statistics and
people analysing them. At this time the neoclassical tradition was not very strong and there
was a whole range of interests. The interest in sociology was shared by quite a few of these
people. Alain Desrosières was already in INSEE, with the same background as I had, together
with another French sociologist, Daniel Berthoux. Both of them, but mainly Alain
Desrosières, was the link to sociology and especially Bourdieu’s sociology. It was the most
sophisticated sociology at that time. It was also well suited to the problem we were dealing
with which was classification.
I was a student of Bourdieu, but not for a very long time. I have even written a perfectly
Bourdieusian paper. Bourdieu was used to rewrite the papers. He said what, for him, was the
very best compliment: “There is not a sentence to change”. So, I learned this language and
this theory. Then I discovered Luc Boltanski, who was actually at this moment departing from
Bourdieu. He had been working with Bourdieu. He was really the main collaborator. I then
began to work with him. This is, of course, a very long story of the construction of a new kind
of sociology, which was completely antagonistic to Bourdieu’s.
One of the problems we have in translating French terms to English is with the concept
dispositifs. It is very difficult to translate. Of course, it was already used by Foucault. For us it
is a very central category. Dispositifs are very interesting in French because it contains
disposé that has the same root as disposition. Disposition is the main category for Bourdieu
2
Department of Social Sciences, Roskilde University, Denmark.
11
because disposition is the habitus. So, in French you are disposé à faire quelque chose
(disposed to do something). Let us begin with that.
Bourdieu worked from the assumption that the disposition is rooted within the person,
incorporated and rooted for all his life. This means that the disposition is the same in all
situations, which is not very dynamical. It is a very good assumption to have for a theory of
reproduction of course, but otherwise you have a rather poor idea of human beings because
they are just developing the same schema all their life in all situations. Actually, we
considered the antagonistic assumption, which is that many dispositions are within the
dispositifs, which means within the arrangement of the situation.
There was also a strong opposition to Bourdieu in terms of politics and morals. For Bourdieu,
of course, it was the Marxist schema of the struggle of social groups. We were thinking that
politics and morals was a big issue for people, that there were many ways to place values on
things. We were working against the way to look at politics just as trial of force and of
collective opposition of groups which is the view by Bourdieu and by many other social
theorists. Actually, we were rebuilding the link between political and moral philosophy.
Sociologists deal with political issues as a question of strength and struggle and with moral
issues as a question of ideology, beliefs and values, which is not really part of their analytical
framework. Even in Weber, who is the most advanced sociologist in this respect, the level of
values is relatively disconnected from the level of rational choice of means.
My first work was on manpower forecasting models which were connected to the French state
planning of education. Since I was building this model I had to work on the connection
between education and occupation. So my very first work was actually on qualification.
Actually the theme of qualification became a sort of a common object of discussion within
this unit of INSEE.
Because of this question of qualification I already have been working on occupational
classification because it was my main tool. The beginning of this idea of plurality came from
the idea that there are deep, deep contrasted ways of what we call in French rapproché, a
word which does not exist in English. There are different ways to bring together, but the idea
of rapproché is more interesting because you have not the idea that at the end they are
together. You have the idea of the process – of getting closer. That means that classification
was not so simple as it was presented because they had somehow to compromise between
these different modes.
It was in 1984 that I published in English the version of Investment in Forms and I was
working together with François Eymard-Duvernay. We developed this notion of qualification
both of people and products. I was working together with François Eymard-Duvernay mainly
on firms and with Luc Boltanski, with whom I began to work at this time, on the qualification
of people and more generally the way they were categorised in the social space. Alain
Desrosières was the one who trained me. We were very close together in the same institution.
André Orléan was working in a very different area, on money. François Eymard-Duvernay
was also in the same institution. It was a very intense milieu where people could work
together and have genuine cooperation.
When I first read your paper on investments in form I read it as a kind of a social
constructivism, but when I was re-reading it I couldn’t really find it. So maybe I just
constructed a social constructivism?
12
I think, somehow, we were social constructivists at that time. Now, I am very opposed to it
because we are considering the fact that there is a process of qualification. What is a social
process? This is just missed in social constructivism. When you say that it is a social
invention, a social process, well you feel satisfied. But for us it was not only a social process.
It was a technical as well as an organisational process, which means that it was not only a
problem of social beliefs and of social representations. It is actually a process that had to be
dealt with as we deal with production processes, but it is much more fundamental. That means
that from the very beginning the notion of investment in form was a project on coordination,
which is not explicitly considered in social constructivism. In social constructivism
coordination is very simple. It is a fact that there is a social group and that people agree on
representations. Coordination is just assumed. For us it was and still is the big issue. I think it
is even more important than convention. Modes of coordination, this is the big issue. So, from
very far you can say that is was a part of the social constructivist turn, but quite rapidly it
departed from social constructivism and opened towards a new program on problematic
uncertain coordination between human beings.
It seems to me that there are three slightly different approaches to describe the plurality of
economic action in the Economics of Convention tradition. Your work with Boltanski on the
theory of grandeur, François Eymard-Duvernay’s notion of conventions of quality and Robert
Salais’ conventions of labour. Can you explain how they are related?
These three are related. The approaches of Olivier Favereau and André Orléan are different.
André Orléan is influenced by René Girard and by the idea of imitation which is in Girard. At
the same time, I think, he has a sort of basic Durkheimian view of the social as a sort of
agreement within a collective. In the case of Olivier Favereau, it is clear too that the
background is different. Maybe the three approaches you mentioned are more or less related
to this notion of a plurality of worth. I prefer worth to greatness. Worth is not a perfect
expression but it focuses on the idea of giving worth which is the main question.
What would you say are the common assumptions of the group of people in the Economics of
Convention tradition?
I think that a common assumption is the idea that coordination is problematic because of
uncertainty. I would say that human beings somehow organize uncertainties. They organize it
to make it make it tractable. For me, this is the answer to Knight’s risk and uncertainty
difference. Risk is the result of channelling uncertainty and making it measurable. But for me
it is a more general issue. What is interesting is the way human beings channel uncertainty in
something that will be testable. Well, of course, it doesn’t work this way, because there is
always new uncertainty. It means that even when you channel, there is, in our sense, a
renewal of the test. And second, because there are many ways to channel uncertainty. Maybe
the notion of convention, in its generality and flexibility, is just a way to point to this
question: How do human beings organise uncertainty so that they have some frame in which
13
they try to coordinate. So again, uncertainty is not at the end of the story. It does not
disappear, but people try to replace it by frames, they tame uncertainty.
This leads us to the idea of shared knowledge which is actually also part of the neoclassical
schema, although they ignore it. We have developed this point in many places. We have tried
to look at shared knowledge as something problematic. It means that the building of
something that will be viewed as common is already problematic, and is already a part of
what we study. That leads also to reconsider the notion of information, and to replace the
unity of information by a view of a variety of formats of information, from very informal
clues to conventionalised formats of information. So uncertainty, the channelling through
common frames to deal with it, which is one way to look at the plurality of worlds, whatever
they are, and the importance of cognition.
I also think we added to that the question of politics and morals. It was not shared among us,
but now I think it is a common preoccupation. It is clear in the recent work by François
Eymard-Duvernay3 and it is also clear in papers by Olivier Favereau who tries to point to the
fact that values have been forgotten by economists. So, now we could add this preoccupation
as a shared one. For me the main problem is the question of genesis of conventions and the
question of what you put within the notion of convention.
Let me give an example. I think that for Robert Salais convention is common expectation
which is actually quite close to common representation. For me it is too demanding and too
close to the classical Durkheimian view of the collective. The reason why I mentioned that is
that you pointed to this question what is common or not in the convention tradition. Strangely,
because somehow I am supposed to be the most inclined sociologist of all of the convention
people, I always criticize the others for being too sociological in the classical meaning of
shared representations and expectations.
Michel Callon argues that the model of economic man could be useful for people engaging in
economic activities. And that the interesting thing about this model is if it is actually used by
economic actors. Do you agree with that argument?
If Callon says that I would say: Why do they use it? I would ask: What kind of properties
should this variety of models have? This is not the kind of question he can answer. He would
just answer that they do use it. I think that the problem with this answer is that it will lack a
reflection on this architecture of regimes and on the path to the public. This is the main
problem for me with this overwhelming notion of network. It doesn’t give any specification of
the link, of the social link, of the social action. And again I think a good specification would
require this specification both of the good and of the reality as it is used as a test. Instead of
that the network modelling in general terms is, I would say, flat, so it cannot give you a good
picture of what is needed to go from proximity to the public and to come back from there.
3
Eymard-Duvernay, F. (2002). "Pour un programme d'économie institutionaliste." Revue Economique
53(2): 325-336.
14
According to François Dosse4 economics of convention is based on three ‘turns’ in the social
sciences: the cognitive turn, the interpretative turn, and the pragmatic turn. Do you agree
with this description?
I think it is rather clear in Dosse’s book because he follows the different other traditions in the
present networks of people who correspond to these different turns. For instance, cognition is
very far from both economics and sociology. In many occasions cognitive science is the
enemy for sociologist and it was just ignored by economists. Because of the creation of
CREA and its director Jean-Pierre Dupuy, cognitive science was imported to France.
I also worked there with André Orléan. It meant that we were concerned with this research on
cognition. I would add that there is a branch of cognition which is not supported at all by this
group at CREA which is social cognition.
The interpretative turn is well known and historically earlier. It was central in connection with
models as a way to grasp. I use grasp. I didn’t use interprete because there might be
drawbacks with interpretation because of this meaning bias. I think the accentuation of the
idea of coordination was a reaction towards a more meaning oriented notion of interpretation.
Of course interpretation comes from the text, hermeneutics. As you know Paul Ricoeur
developed a very strong connection with the theory of action, which makes a much broader
issue of interpretation.
So, we come to the third term which is pragmatics. I could put it this way. We were concerned
with cognition in relation to interpretation, and a need for interpretation which means that we
were not concerned with cognition as reflection about pure brain stories. We were concerned
with cognition in relation to others, which is I think is the core of the interpretation term. And
we were concerned with both cognition and interpretation as far as they were situated in a
pragmatic perspective, in the perspective of acting, which mean not only understanding but
doing things with others. And I think the interest of action is to put to the fore the reality test.
In terms of interpretation, the reality test could be the community of understanding, but we
didn’t want that. We wanted something much more realistic. I would say, I think I said that
maybe last time that for me, and not only for my, I would say and I think Luc Boltanski would
say something similar. We were reacting progressively against social constructivism, which
is, as I said, something we participated somehow in. But very quickly I shifted from social
constructivism, in terms of social representation, to coordination, which means a test of being
social through action. The reason why we are anxious about that is that in many occasions we
are just viewed as social constructivists, which is false.
So now, I think, you see the reason for the three pillars: cognition, interpretation, which is
more than cognition because it is oriented towards the others, and pragmatics, which is more
than interpretation because it is oriented toward action and some return of reality as a reason
of action. But I am sure that Luc Boltanski and several others in the convention tradition will
not say that they are part of the cognitive turn.
4
Dosse, F. (1999). Empire of Meaning: the Humanization of the Social Sciences. Minneapolis,
University of Minnesota Press.
15
How do you see the most important differences between Economics of Convention and
Transaction Cost Economics?
For transaction cost economics I have answered quite extensively in my papers. Let us say
that one can be charitable or one can be very critical. If you are charitable, you will say that
Williamson was the first to try to implement some notion of a plurality of modes of
coordination. He was the first to reflect upon this pluralism. In that sense, we are in the same
move. I don’t say that we do that because of Williamson, because it was really bad and quick.
But the beginning of pluralism is the notion that you have different cognitive forms. So this is
what I can say to be charitable.
Now, if I am critical, and this is a point that we have mentioned many times, we shall say that
he is just missing the point because he, in the end, will consider that all these different modes
of coordination are chosen on the basis of some form of light extension of the classical model
of rational organisation which doesn’t work. If you do that you put at the highest, or more
basic, level these classical implements of economic market coordination, which is exactly
what you wanted to prevent yourself to do through this pluralisation.
How do you see the most important differences between Economics of Convention and
evolutionary institutional theory?
For the evolutionary and institutional current I think the problem which is central in many
theories is that the good is hidden, to say it very shortly. I hate that because I spend so much
effort at doing the contrary. They claim that the good is hidden in something that is called
success. Success without a notion of good is clearly an evaluative notion. You have this
problem already within other evolutionary theories, not specifically the economic current.
Another problem is the notion of routines which again has to be placed in a broader variety of
activities. Why just reduce all to routines? Of course it was a mistake to forget this level. This
is the reason why I tried to analyse it very precisely in the domestic regime. So, it is a good
idea to turn toward this word of behaviour but, I think, the reduction to routines is highly
problematic, because it ignores the main problem of organisation, which is to compose with
this variety of regimes. You cannot make an organisation with routines only, but you cannot
account for the whole functioning of any organisation without this level. I think that the
conceptualisation of the routine level is not appropriate and I think that the conceptualisation
of the test, which is so important for us, in terms of success, is not very adequate either. But
this is just short for a very long story.
How do you see the most important differences between Economics of Convention and
Economic Sociology?
Let us just discuss the seminal paper by Mark Granovetter. Of course, again, one needs to be
very charitable and I will be because it has been so influential on the view on economic
processes as part of more general social processes. Of course I completely agree with the
16
loose notion of embeddedness. In that sense we are all in the same movement. I completely
agree also with the critic of the oversocialised notion of the actor in sociology. This is also
something that I commented a lot on in different papers, specifically considering the
Durkheimian tradition and social constructivism, which has an oversocialised view of the
world.
Considering his critic of the undersocialised actor, I would not agree so much. I think that one
thing that we have done, which is original and which was criticized by most positions, was to
look at market competition as a real, implementable, mode of coordination, under some
conditions of course, which was the qualification of the good, the generalisation of some form
of value, etc. So, I wouldn’t say that the economic actor, in the sense of the one who is acting
on the market, is so undersocialised. He has to have a common knowledge of the good, which
all the other has. And, of course, he has to accept the main form of evaluation, which is,
again, a highly common assumption. So, he is not so atomistic. If you are very precise about
this assumption, then you will say that the economic actor is not alone at all.
In that sense, I think that many people from the economic sociology tradition miss the
understanding of the specificity of market coordination. But of course, market economy is not
economy. That means that they pay attention to many other supports for coordination which
are part of the economic process. So, in that sense they are extremely relevant, useful, and we
are all part of the same story.
One more thing, as I told you before I am not very fund of network theorising as far as
network theories miss the qualification of the links, which is exactly what we try to do. Once
we have done this analysis of the variety of engagements, well then, why not speak of
networks of different engagements? But then network misses its power of unification, which
is the idea that somehow the links are compatible and just allowing the flow of something. I
think this is the problem with network theory. Of course it is extremely efficient in terms of
modelling, but this is the problem that I see from the position of the analysis of the links
themselves. And I would say the same of my friends Michel Callon and Bruno Latour. They
are not in a good position to characterise heterogenity.